UserProof

Evidence-backed synthetic user testing

Monitor your critical product journeys with synthetic users.

UserProof watches signup, onboarding, checkout, and upgrade paths, then turns failures into QA-checked work items with proof, agent prompts, and rerun verification.

Start free trial See how it works

The gap before launch

The bugs that hurt growth rarely look like bugs.

They look like confusion, dead ends, hidden CTAs, mobile traps, and promises the product does not keep yet.

Run the first site check
Pricing intent lost

User chooses Team, signup forgets plan.

URL /pricingStep 04

Conversion risk

CTA below fold
START TRIAL

First fold trace

Console evidence 
POST /signup 422
missing plan_context
networkvisible text
Onboarding dead end

First useful state never appears.

PersonaSkeptical PMGoal: evaluate trust
01 Persona

Goals, objections, device habits.

“I need to know whether this tool is safe to try before I invite a teammate.”

Devicemobile Safari
Browser mission/pricingclick -> signup -> observe
02 Browser mission

Bounded action in a real browser.

Click, type, wait, observe, capture screenshots, collect visible text and failed requests.

Limit20 actions / 10 min
Evidenceseverity highconfidence 0.82
03 Evidence

QA-checked findings with done conditions.

Severity, confidence, repro steps, expected behavior, acceptance criteria, card QA, and rerun notes.

Outputverified work item

How synthetic sessions work

Personas attempt real missions in a real browser.

Bounded sessions collect screenshots, URLs, visible text, console and network evidence, then collapse the trace into findings your team can act on.

See the workflow

Evidence-backed cards

Every finding carries proof and a card QA check.

No vague AI opinions. Screenshots, traces, acceptance criteria, and actionability checks ship together.

High QA checked

Mobile signup loses the selected trial intent

0.82 confidence 3 captures
SCREENSHOT 03 START TRIAL

Observed behavior

After choosing Starter, signup lands on a generic dashboard with no next step or selected plan context.

Expected behavior

The first useful state should preserve the plan, explain the next action, and confirm the trial path.

  1. Open pricing on a 390px mobile viewport.
  2. Tap Starter trial.
  3. Create account.
  4. Dashboard does not preserve plan context.
SCREENSHOT 03 DOM: button[trial] NETWORK: 1 blocked asset QA: testable AC

Kanban plus verification

A card is not verified until the rerun proves it.

Move findings like work. UserProof reruns the mission, verifies passing fixes, and sends failed reruns back to Fixing.

Start a rerun-ready trial
New

Pricing CTA mismatch

Mobile nav hidden

Ready

Onboarding dead end

Signup context lost

Fixing

Trial intent preserved

Verify

Queue rerun

Resolved

Rerun passed

RERUN TRACE Verify -> Resolved
Failure Task Brief

Checkout button remains disabled after validation error

Run
UP-0142
Severity
High
Status
Ready

Includes screenshots, repro steps, acceptance criteria, and rerun instruction.

CodexClaude CodeCursor
{
  "severity": "high",
  "evidence": ["screenshot_03", "dom_excerpt"],
  "source_url": "/pricing",
  "done_when": "CTA is visible above fold",
  "rerun": "repeat checkout mission"
}
shot_01shot_02shot_03
UserProof finds and verifies. Your workflow fixes.

Agent handoff

Send engineering a CLI prompt, not a mystery.

Ready findings include a compact agent prompt for Codex, Claude Code, Cursor, or your team. The prompt tells them to inspect the full evidence through the UserProof CLI before changing code.

Start with agent prompts

Pricing

Choose how many product journeys need coverage.

Plans are sold around journey coverage. Session units stay visible so teams can forecast personas, scheduled runs, and verification reruns.

Capacity example 8 personas x checkout journey = 24 browser session units
Free Trial $0

Prove one public journey with 5 sessions, 1 project, limited evidence, and 7-day retention.

Start
Starter $99/mo

Monitor one public journey with 50 sessions/month, 1 project, 3 seats, and manual runs.

Choose Starter
Team $349/mo

Cover core authenticated journeys with 250 sessions/month, test accounts, scheduled runs, dedupe, agent prompts, and Card QA.

Start Team trial
Growth $899/mo

Run preview and CI regression coverage with 1,000 sessions/month, custom personas, comparisons, and CLI handoff.

Choose Growth
Enterprise Custom

Private runners, custom sessions, SSO, audit logs, redaction, retention, and onboarding.

Start Enterprise setup

Start with one monitored journey

Find friction. QA the card. Rerun after the fix.

Run the first public journey and see what real users would hit next.

Start free trial Read how it works